Surgical and medical treatment of adenomyosis Carl Wood

Human Reproduction Update 1998, Vol. 4, No. 4 pp. 323–336
European Society for Human Reproduction and Embryology
Surgical and medical treatment of adenomyosis
Carl Wood
Department of Obstetrics and Gynaecology, Monash University, Melbourne, Australia
TABLE OF CONTENTS
Introduction
Surgical considerations
Specific surgical techniques
Results of conservative surgery
Medical treatment of adenomyosis
Acknowledgements
References
Surgical considerations
323
323
325
331
332
335
335
The treatment of adenomyosis has been limited by the
difficulty and delay associated with the diagnosis,
often not until after hysterectomy. Magnetic resonance imaging, high resolution vaginal ultrasound and
uterine biopsy have improved early detection of adenomyosis. Drug therapy may be effective in controlling symptoms but the frequent coexistence of
endometriosis and the lack of controlled studies make
their efficacy difficult to quantify. Conservative surgery involving endomyometrial ablation, laparoscopic
myometrial electrocoagulation or excision has proven
to be effective in >50% of patients, although follow-up
has been restricted to 3 years. Hysterectomy will still
be necessary in severe cases of adenomyosis. Early
diagnosis may improve treatment. Investigations are
indicated in women with menstrual pain or menorrhagia not responding to drug therapy.
Key words: adenomyosis/conservative surgery/drug
treatment
Introduction
The oral contraceptive pill, anti-prostaglandins, oral or
parental progestogens, danazol, dimetriose, gonadotrophin
hormone releasing hormone (GnRH) analogues, and
Chinese herbal remedies have all been used to control
menstrual pain and menorrhagia in women with adenomyosis or when drug treatments fail or patients indicate a
preference for surgery.
Diagnosis
Diagnosis of the extent and localization of the disease is
difficult, even with magnetic resonance imaging (MRI),
colour Doppler vaginal ultrasound and uterine biopsy
techniques (Fedele et al., 1992; McCausland, 1992; Popp
et al., 1993; Wood et al., 1993, 1994; Brosens and Barker,
1995; Brosens et al., 1995a; Kang et al., 1996; Reinhold
et al., 1996). MRI is the most sensitive test for detection of
adenomyosis but it is also the most expensive, limiting
access to affluent health care systems (Reinhold et al., 1996).
Uncertainty in defining the site and more particularly the
extent of adenomyosis make it difficult to determine the
feasibility and accuracy of complete excision when
conserving the uterus. This is one reason why hysterectomy
has remained the most popular operation for adenomyosis.
Extent of removal
One factor which may favour attempts at excision of
localized areas of adenomyosis is the possibility of reduction
or cure of symptoms even when excision is incomplete.
Incomplete electrocoagulation or excision in patients has
resulted in symptom relief for 3 years (C.Wood, unpublished
data).
Technical difficulty in excision
Laparoscopic surgery may be limited by the need to excise
ill-defined, tough adenomyotic tissue and to use robust
suturing equipment to obtain wound closure after excising
significant areas of myometrium. The easier removal of
adenomyosis by laparotomy is a less attractive alternative to
laparoscopy, particularly as cure cannot be guaranteed.
Unless the adenomyosis is well defined, as in an
adenomyoma, it is not possible to be certain of cure
following excision or electrocoagulation.
Pregnancy after myometrial excision and
coagulation
One problem when conserving the uterus is the uncertainty of
the extent of myometrium which can be removed and still
allow normal pregnancy and birth. Comparison with the
324
C.Wood
relative safety of pregnancy and birth after myomectomy may
be inappropriate. Fibroids grow inside normal myometrium
which they displace as in a benign tumour. When the fibroids
are removed, a capsule demarcates them from normal
myometrium. The cavity is then repaired leaving the uterus
similar to what it was prior to growth of the fibroid, except for
the presence of a scar. Adenomyosis infiltrates normal
myometrium so that excision of the diseased area subtracts
myometrial mass from the total uterine volume. Removal of
significant amounts of myometrium poses two problems, (i) a
reduction in myometrial capacity of the uterus during
pregnancy, which may predispose to abortion or premature
labour, and (ii) the production of uterine scars, which may
contain foci of undetectable adenomyosis and may have
reduced tensile strength.
Wound apposition may be more difficult to achieve after
excision of adenomyosis as the loss of circumference of the
myometrium will increase tension in the stretched
myometrium when it is opposed to fill the gap. When
one-third of the posterior myometrium was excised in a
patient with adenomyosis, a size one vicryl stitch placed in
two layers at laparotomy was unable to oppose the
myometrial edges, which was only achieved by a third suture
layer. Uterine relaxants, for example, β-mimetics, may assist
apposition. A curved suture line may be more effective as the
tension per linear centimetre would be reduced by the longer
curved margin of the incision, allowing easier apposition.
The increased expansile capacity of the uterus in
pregnancy depends mostly on an increase in plasticity rather
than elasticity (Zimmer, 1962; Wood, 1964). The
rearrangement of spiral bundles of muscle from a transverse
to longitudinal arrangement in pregnancy does not account
for the change in plasticity (Goerttler, 1930). The main factor
determining the increase in plasticity is the collagenous
framework of connective tissue (Harkness and Harkness,
1959). Microfibrils of collagen are individually relatively
inexpansible, and if arranged in a long axis cannot be
extended more than ~10% before rupture (Stucke, 1950). In
tissues such as the uterus the microfibrils of collagen are
arranged in a three-dimensional network. Initially the fibres
are straightened, then there is slip between microfibrils
increasing plasticity and uterine expansion until the end of
pregnancy. Adenomyosis may reduce the expansile capacity
of the uterus in two ways, by replacing normal myometrium
and connective tissue with adenomyotic tissue, which may
distort both the spiral arrangement of muscle fibres and the
three-dimensional network of collagen, or by adding scar
tissue subsequent to excision of the adenomyosis. In a study
of spontaneous uterine rupture in pregnancy adenomyosis
was found at the site of the uterine rupture in three cases
(Wood, 1960). The scar tissue following excision is a
separate risk factor. Coagulation of adenomyosis will have a
similar effect to excision as myometrial mass is reduced and
scar tissue is formed. The scar may be more extensive after
coagulation as abnormal tissue is not removed.
The increased expansile properties of uterine myometrium
and connective tissue in pregnancy may allow excision of a
considerable volume of myometrium without preventing
normal uterine expansion in pregnancy. The largest area of
myometrium removed in our own experience is one-half of
the posterior uterine myometrium. The woman conceived
and a normal baby was delivered by elective Caesarean
section at 37 weeks of pregnancy. The experience of women
with pregnancy in a unicornuate uterus may be relevant. The
uterus formed from one Müllerian duct is smaller than
normal and although premature labour is more frequent
(15%) a normal duration of pregnancy is most common.
Myometrial mass may be less important than the myometrial
integrity of the myometrium and connective tissue.
Pregnancy following excision or coagulation of
adenomyosis poses special problems which have not been
resolved. Documentation of the extent of the surgical
procedures and pregnancy outcomes is necessary so that
women may be better informed of the possible risks.
Early detection may favour conservative surgery?
In subfertile women with menorrhagia and dysmenorrhoea,
adenomyosis has been found in 28 of 56 women having MRI
and uterine histology (Brosens et al., 1995b). MRI diagnosis
was based on the presence of a distorted endomyometrial
junction (EMJ) and severe menorrhagia was more common in
the women with adenomyosis. Such a high frequency of
menorrhagia supports the concept that the disease may be
underdiagnosed as it is usually recognized only when
myometrial thickening and distortion of the surface contour of
the uterus are present at the time of surgery, and by histology
after hysterectomy. If the only investigative manifestation of
adenomyosis is distortion of the EMJ, the possibility of
conservative surgery is limited by the need to perform MRI to
make the diagnosis and by limitation on the type of surgery.
Endomyometrial ablation may be effective in reducing
menorrhagia but would be limited to women not wishing to
conceive (Popp et al., 1993; Wood et al., 1994). If the EMJ
was distorted over a small area it may be possible to perform a
limited endomyometrial ablation and thus conserve fertility. A
24 week pregnancy has been reported after planned subfundal
endometrial ablation (Wood and Rogers, 1993) and also after
endometrial ablation when significant areas of endometrium
have been unintentionally left intact. There is uncertainty as to
how large an area of endometrium is required to sustain a
normal term pregnancy. The variety of clinical situations
Treatment of adenomyosis
where endometrium is partially damaged and subsequent full
term pregnancy has been recorded, e.g. endometrial ablation,
removal of uterine synechiae, and resection of large
submucous myomata, suggests that at least one-third of the
endometrium may be removed without compromising normal
placentation. The placenta normally occupies <50% of the
endometrial surface.
Specific surgical techniques
The choice of a suitable surgical procedure depends upon the
site and extent of disease, the age of the patient, the desire for
future pregnancy, the patient’s desire for certain cure or not,
and the surgical skill of the gynaecologist.
Endomyometrial ablation/resection
Endomyometrial resection is most suited to patients with
disease limited to the EMJ as menstrual symptoms may be
reduced and the pathology may be removed. It may also be
useful when adenomyosis is present in the outer myometrium as laparoscopic myometrial excision alone may not
cure menstrual symptoms, either because excision may be
incomplete or the menstrual symptoms are not caused by the
outer myometrial adenomyosis. Desire for a future
pregnancy contraindicates endomyometrial resection (see
above). Adenomyosis has been found incidentally in seven
of 29 women on hormone replacement therapy having endometrial ablation for menopausal bleeding (Phillips, 1995).
Technique
The technique of endometrial ablation has been well
described. If MRI or ultrasound shows the extent and site of
endomyometrial distortion the procedure can be modified to
include 2–3 mm of myometrium in the affected areas. The
whole of the endometrium should be removed as menorrhagia
may be due to factors other than the adenomyosis. Deeper
myometrial removal or ablation carries the risk of causing
increased bleeding as significant arteries are situated ~5 mm
deep to the myometrial surface. Histology of the excised
myometrial fragments may help to confirm or refute the
diagnosis.
When endomyometrial resection has been performed as a
single operative procedure, menstrual symptoms have been
controlled in 55% of women for at least 2 years (Table I).
Laparoscopic myometrial electrocoagulation
Electrocoagulation has the capability of shrinking
adenomyosis by causing necrosis. The technique has been
applied to localized or extensive disease. The adenomyosis
can be detected by MRI, vaginal ultrasound, inspection of the
uterus at laparoscopy, myometrial needling, or manual
palpation during gasless laparoscopy to detect differences in
consistency between normal and abnormal tissue. Electrocoagulation may be less accurate than surgical excision as
electrical conduction in the abnormal tissue may be
incomplete and this cannot be checked at the time of surgery.
It may also reduce the strength of the myometrium by
replacing abnormal myometrium with scar tissue. The width
of the scar may be more extensive than after surgical excision
when close apposition of normal myometrium is achieved.
Extensive myometrial electrocoagulation has been
performed in two women aged 42 and 46 years with
extensive adenomyosis on the anterior and posterior uterine
walls; drug therapy had failed, excision was not feasible, and
hysterectomy was not wanted. Two years later both are free
of severe menstrual pain and bleeding. Diffuse multifocal
electrocoagulation of the myometrium containing adenomyosis may be sufficient to control symptoms. The risk of
uterine rupture following extensive electrocoagulation is
demonstrated by the following experience. One patient had
two laparoscopic procedures involving myometrial electrocoagulation, one of which was also associated with excision
of an elevated adenomyotic area. The patient was aware of
the risk of uterine rupture, she had not responded to GnRH
analogue therapy, was not suitable for extensive myometrial
excision and had refused hysterectomy as she wished to
attempt conception even if this failed. A subsequent
pregnancy resulted in uterine rupture at 12 weeks.
Table I. Results of conservative surgery, 1991–1997
Symptom-free
Endomyometrial
resection
Myometrial
electrocoagulation
Myometrial
excision
Subsequent
hysterectomy
aThree
Hospital stay
Complications
n
6 months
24 months
(days)
18
12
10 (55%)
1
0
11
7
6 (55%)
1–2
0
25
20
16 (64%)
1–3
3a
patients had a temperature of >37.5°C for >2 days.
325
10/54
(18.5%)
326
C.Wood
Figure 1. Bipolar needle used to electrocoagulate outer myometrial
adenomyosis. Surface charring is avoided by restricting coagulation
close to the serosal surface.
Electrocoagulation is best suited to women over 40 years
of age, who do not wish to conceive, and who wish to avoid
more extensive surgery such as excision or hysterectomy.
Even following recurrence the procedure may be repeated
until the onset of the menopause when symptoms cease.
Technique
Uterine manipulation with a Valtchev manipulator improves
access to the diseased areas by facilitating antero-posterior
and lateral movement of the uterus.
Vasoconstricting agents such as adrenaline and vasopressin are not used routinely as excessive bleeding has not
been experienced and the blanching of the myometrium after
vasoconstriction makes it difficult to determine the
devascularizing effect of electro-coagulation or uterine
vessel closure.
Closure of the ascending uterine artery may be performed
if technically feasible, future pregnancy is not wanted, and
the site of the adenomyosis is in the upper uterine body.
Bipolar forceps, clips or suture ligation may be used to close
the uterine vessels.
Electrocoagulation of the adenomyosis may be carried out
with unipolar or bipolar needles, using 50 W coagulation
current (Figure 1). Bipolar needles have a theoretical
advantage of concentrating current between the two needles,
but their effectiveness is diminished by the tendency of the
two needles to move close together as they penetrate the
myometrium. Additionally, the area of coagulation may
spread outwards from each needle, simulating the effect of
monopolar electrocoagulation.
The extent of coagulation can be controlled by reducing
the current strength and changing the time the needle(s) are
held in position. In order to reduce the possibility of severe
surface necrosis and carbonization, either of which may
encourage future adhesion formation, the insulated part of
the needle is buried a few millimetres below the uterine
surface before electrocoagulation is commenced. The
insulation on the bipolar needle can be extended so that the
active part of the electrode is shortened in order to avoid
surface coagulation and necrosis. Needle punctures are made
at 1–2 cm intervals, depending on the spread of the
coagulative effect. The depth of needle puncture may vary,
depending on the thickness of the adenomyotic myometrium
determined preoperatively by ultrasound or MRI. This varies
from 3 to 25 mm. If hysteroscopic endomyometrial ablation
has also been carried out, the depth of laparoscopic needle
electrocoagulation may be reduced. Lasers have been used to
shrink fibroids but their use has not been reported in the
laparoscopic treatment of adenomyosis.
Hysteroscopic endomyometrial ablation may be
performed in association with myometrial electrocoagulation as menorrhagia and dysmenorrhoea may not be
related to the presence of outer myometrial adenomyosis.
Distortion of the EMJ is probably one of the causes of the
menorrhagia and dysmenorrhoea in adenomyosis.
We have not been able to visualize the depth of
laparoscopic myometrial coagulation by hysteroscopic
inspection of the uterine cavity during the operation.
Bleeding is rare during electrocoagulation and can be
controlled by using a vasopressor or myotonic drugs such as
adrenaline, oxytoxin or vasopressin, or by bipolar
electro-coagulation or suture ligation. Patients are usually in
hospital for 8–24 h. No complications have been observed
including post-operative infection, bleeding or subsequent
adhesion formation.
The result of the surgery may be assessed by symptom
relief and MRI or vaginal ultrasound. Loss of features of
adenomyosis including reduction of myometrial thickness,
reduced vascularity and normal myometrial appearance have
all been observed. Symptom relief may occur and persist for
several years in the presence of reduced ultrasound evidence
of adenomyosis. Sterilization should be offered to all women
having myometrial electrocoagulation because of the
possible future risk of uterine rupture in pregnancy.
Myometrial excision
Adenomyosis may be excised if it does not involve the major
portion of the uterus, and its extent can be defined as
previously described. The technique is also suitable for
adenomyomas where the margins of the pathology are more
easily defined. It may be useful in women wishing to become
pregnant, providing sufficient myometrium remains to allow
uterine expansion and term pregnancy and the scar formed
Treatment of adenomyosis
after excision is not wide or shallow. MRI or colour Doppler
ultrasound after surgery should be used to check both for
cure, the width and depth of scar, and the possible association
of residual adenomyosis close to the scar, before attempts at
conception are advised.
Technique
Preoperative GnRH analogues or danazol may reduce
uterine vascularity, correct anaemia if the patient has severe
menorrhagia, and reduce operative bleeding which
facilitates surgery by laparoscopy rather than laparotomy.
Vasoconstrictor drugs may also reduce bleeding at the time
of surgery.
Prior to myometrial excision, as with electrosurgical
coagulation, the uterine blood supply may be reduced by
suture or clip ligation or bipolar diathermy of the ascending
uterine vessels in women not concerned with fertility. Apart
from reducing bleeding during surgery the reduction in
blood flow may reduce future growth or development of
adenomyosis.
Two associated surgical procedures may be offered:
sterilization to prevent conception, and hysteroscopic
endomyometrial ablation if menorrhagia is present and
fertility is not required.
Laparoscopy, and gasless laparoscopy, with or without
minilaparotomy, facilitate myometrial excision avoiding the
need to perform laparotomy. Gasless laparoscopy is done
with a Maher abdominal elevator, forming an S-shaped loop;
this is effective and cheap (Maher, 1995; Wood and Maher,
1996a). A finger or laparotomy instruments can gain entry to
the abdomen through a 2–4 cm incision which may be
sufficient to remove and repair areas of myometrium up to
6×8 cm.
A Valtchev uterine manipulator is used to position the
adenomyotic areas as close as possible to a laparoscopic or
minilaparotomy incision. Sometimes a myoma screw may
stabilize the diseased area and aid excision. A diathermy
spoon using 100 W monopolar current, or scalpel, is suitable
for excision. The spoon has the advantage of cutting
effectively with the sharp end close to the tissue, and of
coagulating vessels when the convex curve of the spoon
compresses the vessel. When the tissue is very firm the
scalpel may be preferable, providing more effective and
rapid excision. The scalpel can be used safely through a 2 cm
accessory laparoscopy incision or a minilaparotomy
incision. The margin of the adenomyosis may be determined
by change in appearance, vascularity or consistency; finger
palpation may be an advantage.
A myometrial morcellator may also be used to remove
adenomyotic tissue, coring pieces up to 15–20 mm in
327
diameter. The difficulty in defining the margin of
adenomyotic tissue makes morcellation less precise than
scissor or knife dissection. The morcellator hides the tissue
as it is cored out. The risk of trauma to other organs is
prevented by the myometrium being drawn outwards or by
the instrument not being inserted beyond the surface of the
uterus. The morcellator may be hand- or electrically driven.
It costs Aus$7000–12 000. Lateral insertion in the
abdominal wall is essential for safety. A 10 mm laparoscope
gives a better view of the procedure. Laparoscopy or gasless
laparoscopy, using a large scalpel blade and/or large heavy
scissors to morcellate the fibroid as it is withdrawn from a
small 2–4 cm incision in the umbilicus, the suprapubic area
or vagina, enables removal of fibroids up to 1000 g (Pelosi
and Kadar, 1994; Wood and Maher, 1996b). This technique
is cheaper than a morcellator, is just as quick, and may be
safer as the surgery is done under vision in the abdominal
wound.
Closure of incisions longer than 5–6 cm may require
laparotomy instruments as excision of a significant volume
of myometrium increases the tension at the myometrial
edges which may have to be stretched to close the defect. If
the uterine wound is brought into a minilaparotomy incision,
the defect can be closed more easily and quickly. Absorbable
sutures (No. 1) are used in one or more layers. If there is a
large defect a single layer through-and-through suture may
best approximate the wound, acting as a tension suture, and
because of the increased thickness of the whole myometrium
it is less likely to tear as tension is increased to attain closure.
Anti-adhesives such as Interceed and Goretex
membrane may be used (Diamond et al., 1987; Jansen, 1991;
Operative Laparoscopy Study Group, 1991; Bulletti et al.,
1996) (Figure 2). The frequency of adhesions after excision
of adenomyosis has not been reported. Interceed may be
used if perfect haemostasis is obtained. Application of
Surgicel prior to Interceed may improve haemostasis and
allow the use of Interceed (Figure 3). If bleeding persists
Goretex can be stapled over the wound. This need not be
removed unless pregnancy is planned. Uterine enlargement
may displace the membrane from the uterus which may
attach to other organs. Physiological solutions have been
used to reduce adhesive formation but their efficacy in
animal trials has been less than Interceed or Goretex (Jansen,
1991).
The use and safety of myometrial excision may be
assessed by comparison to the results of laparoscopic
myomectomy. Fibroids up to 1000 g have been removed
laparoscopically by modifying surgical techniques (Pelosi
and Kadar, 1994; Wood and Maher, 1996b). The safety of
the laparoscopic technique has been established in 346
patients in four reports, one postoperative haemorrhage
328
C.Wood
Partial or subtotal hysterectomy
Figure 2. Goretex membrane stapled over two-thirds of the posterior
uterine surface to prevent adhesion formation.
Myometrial excision can be extended to remove the major
portion of the uterus. Women who do not wish to conceive
may still prefer to retain the normal part of the uterus, the
reasons being: that only diseased uterine tissue need be
removed; that retaining the uterus is emotionally important;
and that possible complications of hysterectomy, increased
operative morbidity, and an earlier menopause may be
avoided. In one woman aged 46 years, with a uterus enlarged
to the size of a 20 week pregnancy by adenomyosis, 80% of
the uterus was removed in association with partial
endometrial ablation, and the women has had painless scanty
menses for >5 years.
Laparoscopic partial or subtotal hysterectomy
Figure 3. Surgicel is placed over an extensive myometrial wound
following excision of adenomyosis when there was difficulty obtaining haemostasis. This led to wound closure with good haemostasis. Interceed was then placed over a dry suture line.
requiring reoperation being the only serious complication
(Darai et al., 1997; Pelosi and Pelosi, 1997; Seinera et al.,
1997; Wood and Maher, 1997). Conversion to laparotomy
only occurred in two of the five studies, the incidence being
1% and 28%. The average hospital stay for abdominal
myomectomy is 4–5 days, whereas the laparoscopically
associated procedure is 1–3 days (Hirsch, 1993). The cost of
laparoscopic myomectomy (Aus$2217) has been estimated
to be lower than abdominal myomectomy (Aus$3825) due
to reduced hospital costs (Hirsch, 1993). The reduction in
hospital stay, with the probable associated advantages of
reduced pain, reduced risk of wound complications, earlier
return to normal activity, and reduced costs, also suggests it is
a suitable alternative to abdominal myomectomy.
When adenomyosis is extensive, involving more than
one-third of the uterus, and fertility is not required, it may be
easier to remove the top half or two-thirds of the uterus
leaving any normal myometrium. The residual endometrium
can easily be removed if menstruation is not wanted.
Subtotal hysterectomy is preferable to partial hysterectomy
in extensive adenomyosis as recurrence of adenomyosis is
less likely and cure of menorrhagia and dysmenorrhoea
more certain. Disadvantages of subtotal compared to total
hysterectomy are: a small risk of residual or recurrent
adenomyosis in the cervix; difficulty in removal of
associated rectovaginal adenomyosis or adenomyoma,
particularly if attached to the cervix; and the possibility of
cervical abnormalities developing which may require further
surgery. The possible advantages of subtotal compared to
total hysterectomy include shorter operating time, reduced
blood loss, earlier patient discharge from hospital and return
to normal activity, reduced risk of bladder and ureteric
trauma and reduced risk of adverse effects on sexual and
bladder function (Wood and Maher, 1997). Controlled trials
comparing total and subtotal hysterectomy are inconclusive,
because they have mainly dealt with differences in operative
and early postoperative events, are too few in number to
determine if bladder and ureteric trauma is different, have
not shown significant differences in sexual function, and the
surgeons involved have not demonstrated comparable
efficacy in the two surgical techniques prior to embarking on
the trial. If the surgeon finds laparoscopic total hysterectomy
difficult because of inexperience, difficult pathology or
technical surgical problems, subtotal hysterectomy is
certainly preferable. Further multicentre, carefully controlled
studies and follow-up of patients for one year by surgeons
equally experienced in both techniques is required.
Treatment of adenomyosis
329
Total hysterectomy
Hysterectomy is the most common operation for
adenomyosis as it nearly always ensures cure, and avoids
both difficulty in defining the extent of the disease, a
requirement for successful conservative surgery, and
technical surgical difficulties, which may be associated with
myometrial excision or electrocoagulation. Hysterectomy
has the disadvantage of being associated with ureteric, bowel
and bladder trauma in 1–2% of patients (Wood and Maher,
1997), and prolonged hospital stay and return to normal
activity when compared to conservative surgery (Table I).
The failure of conservative surgery may result in delayed
hysterectomy in at least 10% of patients (Table I)
Figure 4. Adenomyosis of anterior uterine wall and extension onto
bladder surface with infiltration into muscular layer of bladder.
Vaginal or laparoscopic total hysterectomy
Providing no pelvic endometriosis is present, which can be
determined by preoperative or operative laparoscopy,
vaginal hysterectomy may be the procedure of choice.
Controlled trials have shown it to be equally or more
effective than laparoscopic hysterectomy, operating time
being shorter and costs perhaps lower (Wood and Maher,
1997). A review of 70 articles concerning various types of
hysterectomy shows that bleeding, use of blood transfusion
and unexplained fever are significantly more common after
vaginal than laparoscopic hysterectomy (Wood and Maher,
1997). Providing a check laparoscopy is performed at the
completion of vaginal hysterectomy, and any bleeding
detected and corrected, it may still be more cost effective
than the laparoscopic procedure (Wood and Maher, 1997).
As surgeons vary in their ability to perform vaginal
hysterectomy—from 20 to 90% of hysterectomies are
performed by this technique—laparoscopic hysterectomy
also has an important role in performance of hysterectomy
for adenomyosis (Wood and Maher, 1997). It has the
advantage of enabling detection and removal of associated
endometriosis, which is not possible during vaginal
hysterectomy, and to deal more easily with very large uteri
>500 g and associated adnexal pathology (Wood and Maher,
1997).
Abdominal hysterectomy is associated with increased
costs, longer stay in hospital, 2–4 days, and delayed return to
normal activities, when compared to laparoscopic
hysterectomy in controlled trials (Wood and Maher, 1997).
The incidence of trauma to the bladder, ureter and bowel are
not different between abdominal, laparoscopic and vaginal
hysterectomy (Wood and Maher, 1997). Laparotomy is
required if the surgeon is not skilled in the vaginal or
laparoscopic technique, if complications occur which may
require laparotomy to repair trauma, or if associated
pathology such as large fibroids or severe adhesions are
present.
In order to completely remove uterine adenomyosis,
surgery may need to be extended into the rectovaginal
septum or bladder, when adenomyosis is either associated
with, or has extended from, uterine adenomyosis.
Arterial embolization
Reduction of uterine blood flow by arterial embolization has
been shown to reduce the growth of fibroids, (Ravina et al.,
1995) and may be applicable to the treatment of
adenomyosis.
Extrauterine adenomyosis
This has been reported in the broad ligament, the bladder and
rectovaginal septum (Figure 4). The one patient with
adenomyosis in the broad ligament was on tamoxifen after
treatment for breast carcinoma (Chung et al., 1997). A
laparotomy and hysterectomy and bilateral salpingooophorectomy was performed and revealed adenomyosis
with cyst formation and a thick capsule.
Lesions in the rectovaginal septum have been confused
with endometriosis (Donnez and Nissole, 1995). The
response to drug therapy is usually ineffectual or
incomplete. Surgical removal can be achieved by
laparoscopy or laparovaginal surgery with an abdominal
elevator (Maher, 1995; Maher et al., 1995; Wood and Maher,
1996a). A bowel preparation, rectal probe and uterine
manipulator assist inspection of the pouch of Douglas.
Adhesions obscuring the nodule are cleared. Access to the
nodule may be assisted by ureteric dissection, surgical
dissection in the pararectal space, and finger or probe display
of the vaginal vault. Once the rectovaginal space is defined,
330
C.Wood
the further dissection of the infiltrating lesion is
straightforward. Nodules up to 5 or 6 cm have been excised.
Although laser, diathermy or scissors can be used to excise
the nodule, thermal ablation alone is inadequate as only the
superficial portion of the nodule is dealt with.
If the vagina is likely to be opened, placement of an
abdominal elevator will allow continued dissection after
vaginal opening. In the absence of an elevator, opening of the
vagina can be left to the end of the procedure, the nodule
being passed into the vagina for final removal.
Adenomyosis on the anterior uterine wall may extend onto
the surface of the bladder. This can be removed with the
uterine nodule (Figure 4). A cystoscopy excludes extension
of disease into the bladder. A metal catheter in the bladder
assists identification of the margin of bladder muscle during
dissection. If the nodule is >1–2 cm diameter, placement of
ureteric catheters reduces the risk of ureteric trauma if
dissection is required low and lateral on the anterior uterine
surface.
Fibroids and adenomyosis
Adenomyosis has been found in 23% of uteri removed
because of the presence of fibroids (Vercellini et al., 1995).
Because they are both common conditions, they often
coexist particularly in larger uteri (Lev Gur, 1996). Because
conservative treatment for fibroids is also feasible by
electrocoagulation
(myolysis),
laparoscopic
or
minilaparotomy excision, and ivalon particle artery
embolization (Chung et al., 1997). the treatment is unlikely
to be different. It is more difficult to diagnose adenomyosis
in the presence of fibroids, so that preoperative counselling
of the patient may be incomplete. During surgery a fibroid
may be easily enucleated and then an adenomyotic nodule
without a clear plane of enucleation encountered. It may be
difficult to distinguish adenomyosis from a degenerate
fibroid or from a fibroid that has adhered to the capsule
following necrosis after GnRH analogue treatment. Nodules
without a distinct capsule may be removed with a margin of
surrounding myometrium to allow for the possibility of the
nodule being adenomyosis.
In one patient with both multiple fibroids and
adenomyosis, hemihysterectomy was performed, as the
patient wished to retain any normal uterus and to menstruate.
The cervix and 2 cm of the lower uterine body were
conserved. The patient remains symptom-free after 5 years.
If both conditions are diagnosed preoperatively, the results
of conservative surgery are most likely to mimic that of
adenomyosis, with a higher failure rate for symptom
removal. The patient may favour hysterectomy as a more
certain cure.
Adenomyosis and uterine cancer
Precancerous changes, utilizing monoclonal antibodies
against P53, were studied in 56 women with endometrial
cancer associated with adenomyosis (Taskin et al., 1996).
Ten women without endometrial cancer and adenomyosis
were a control group. This showed that the precancerous
changes in adenomyosis are most likely due to a carcinogenic field effect in the vicinity of the endometrial cancer
rather than direct invasion. This makes it less likely that
adenomyosis is precancerous any more than the uterine
endometrium. The frequency of adenomyotic endometrial
cancer in the presence of adenomyosis and normal uterine
endometrium is unknown.
Adenocarcinoma within adenomyosis may have a better
prognosis than adeno-carcinoma invading the myometrium
(Mittal and Barwick, 1993). There were no deaths in 18
cases of adenomyosis cancer and eight deaths in 43 cases
(19%) of adenocarcinoma invading the myometrium. The
adenocarcinomas in adenomyosis were characterized by
frequent preceding oestrogen use, low histological grades
and a good prognosis. It is possible that cessation of
oestrogen and treatment with medroxyprogesterone acetate
may reverse the histology and avoid hysterectomy and
oophorectomy in such patients. Hysterectomy may still be
preferable if the oestrogen is making a contribution to the
patient’s quality of life or future physical health.
It may be prudent to advise women desiring conservative
surgery for adenomyosis to consider hysterectomy if they
have an increased risk of uterine cancer, e.g. obesity,
diabetes, polycystic ovaries or a family history.
In postmenopausal women treated for breast cancer and
on tamoxifen, adenomyosis has been found more frequently,
affecting 14 (8%) of 173 women, which is 3–4-fold higher
than the expected incidence (Cohen et al., 1995). These
women had hysterectomy and oophorectomy for bleeding
symptoms. In order to avoid adenomyosis and surgery, low
dose progestogens may be worthwhile, providing they do
not adversely affect the anticancer effect of tamoxifen.
Adenomyosis following endometrial ablation
Adenomyosis has been reported to follow endometrial
ablation by use of resection or the rollerball. In a report of
two patients ablation was performed in the apparent absence
of uterine pathology. The endometrium was not prepared.
Adenomyosis was discovered after hysterectomy performed
because of recurrence of menorrhagia and dysmenorrhoea
(Yuen, 1995).
It is difficult to prove that endometrial ablation or resection
causes adenomyosis, because even myometrial biopsy prior
to or at the time of resection may be negative in the presence
Treatment of adenomyosis
of adenomyosis. It is possible that disorganization of the
endomyometrial interface may occur at the time of surgery,
removing normal physiological controls that prevent
endometrial penetration of the myometrium. The situation
may be analogous to placentation when penetration of
myometrium may occur in some circumstances and lead to
pathological attachment inside the myometrium.
There is circumstantial evidence in case or anecdotal
reports that endometrial ablation/resection may have caused
adenomyosis where menorrhagia was present before
surgery, where menstrual pain and uterine enlargement
followed surgery, and where adenomyosis was subsequently
proven by histology. The most common explanation for
failure of endometrial ablation/resection focuses on the
failure to remove sufficient endometrium at the time of
surgery and/or subsequent regrowth of the endometrium.
Other failures may result from the late development of
adenomyosis, or from adenomyosis present at the time of
surgery. Checks for the presence of adenomyosis may not
have been performed prior to endometrial ablation/resection,
and even if they have, ultrasound, MRI and myometrial
biopsy may give false negative results. In 42 women having
routine myometrial biopsy at the time of endometrial
ablation/resection, seven were shown to have adenomyosis
(Wood, 1992).
Despite uncertainty, the possibility of a causal link
between endometrial ablation/resection and the subsequent
development of adenomyosis remains. Pre-preoperative
thinning of the endometrium by GnRH analogues, complete
endometrial removal, and postoperative suppression of
endometrial growth during healing, may reduce the risk of
operative or postoperative endometrial penetration and
survival in the myometrium.
Results of conservative surgery
The results of conservative surgery in a personal series of
women with adenomyosis diagnosed by vaginal ultrasound
with vascular assessment, percutaneous uterine biopsy, and
histology of excised endomyometrial or myometrial
fragments at the time of surgery, are presented in Tables I, II
and III. Sixty three per cent of women were symptom-free 2
years later and 12% required hysterectomy during the same
time period because of persistence or recurrence of severe
symptoms. Each of the techniques had a success rate >50%.
Nine of 16 women attempting pregnancy conceived, four of
seven after myometrial electrocoagulation and five of nine
after myometrial excision. One woman who had two
electrocoagulation treatments, including one associated
myometrial excision, ruptured her uterus in the twelfth week
of pregnancy.
331
Table II. Indications for conservative surgery for adenomyosis
n = 54
46
Indications for surgery
Failed medical treatment
Oral contraceptive pill
30
Progestogens
8
Danocrine
3
GnRH analogue
5
Patient preference for surgical treatment
8
GnRH = gonadotrophin releasing hormone.
Table III. Clinical features of patients having conservative surgery for adenomyosis (n = 54)
Age (years)
24–52
Parity
0–5
Menorrhagia
45
Dysmenorrhoea
48
Dyspareunia
15
Infertility
12
Site of adenomyosis determined by biopsy or surgical histology:
inner myometrium, 18 patients; outer myometrium, 36 patients.
Outer myometrium had normal appearance at laparoscopy and
vaginal ultrasound examination with colour Doppler.
Adenomyosis also present in inner myometrium in 17 patients.
MRI and uterine biopsy were used to diagnose nodular
adenomyosis by Phillips et al. (1996). Preoperative GnRH
analogue, endomyometrial resection and bipolar coagulation
were used in 14 women. One year after treatment
menorrhagia was cured in 12 and dysmenorrhoea in eight.
Two proceeded to hysterectomy. The advantage of
preoperative use of GnRH analogue was shown by a 50.8%
mean reduction of uterine volume after leuprolide acetate
treatment for 3 months. A further 14.9% mean reduction
occurred after the surgery. The beneficial effect of leuprolide
may have continued for at least 3 months after surgery so that
a 1 year follow-up assessment may give a favourable view of
the efficacy of the surgery. In 10 patients laparoscopic
bipolar coagulation alone was performed on adenomyomata,
and seven of 10 women had symptom relief after 1 year, one
requiring hysterectomy (Phillips et al., 1996). In a 2 year
follow-up (Table I) the results were similar following
electrocoagulation without the use of a GnRH analogue.
Endomyometrial ablation alone may be successful in
curing symptoms of menorrhagia and dysmenorrhoea (Table
I; Phillips et al., 1996). Using rollerball ablation and
performing a posterior uterine wall biopsy prior to this to
determine the depth of penetration shows that those with
minimal penetration had a good outcome and those with
deep penetration a poor outcome. The authors recommended
hysterectomy in the presence of deep penetration as a
rollerball procedure would only cause necrosis in the
superficial 2–3 mm of the myometrium. An electrical loop
332
C.Wood
may be used to remove >2–3 mm of myometrium, although
the risk of bleeding is increased. Laparoscopic myometrial
electrocoagulation may be useful in such circumstances as
penetration of coagulation may be achieved over the full
depth of the myometrium. One of the difficulties assessing
results of endomyometrial ablation is the lack of certainty
of diagnosis, even with histology, because of false negatives, and uncertainty of the MRI criteria for diagnosis. The
specificity of the diagnosis by MRI has been determined by
a junctional zone thicker than 5 mm. Thickness >5 mm has
been found in 40% of normal subjects having serial MRI
measurements which also showed thickening up to 12 mm
and focal myometrial bulging which may result from uterine contractions (Kang et al., 1996).
The diagnosis of adenomyosis and the assessment of
surgical procedures has been complicated further by the
hypothesis that adenomyosis is a dichotomous disease
characterized primarily by the disruption of the inner
myometrium (junctional zone hypertrophy) and its function,
with secondary infiltration of endometrium into the
myometrium under certain circumstances (Brosens et al.,
1995b). The former may exist without the latter and lead to
menorrhagia but not menstrual pain. Proliferation of the
inner myometrium may result from endometrial or immune
factors and medical treatment may become more appropriate
than surgical removal of the endometrium, particularly as it
is more common in young women with menorrhagia
(Brosens et al., 1995b). Both choice of treatment and
surgical results may need to be classified by the results of
MRI, although the uncertainty of accuracy in detecting both
junctional zone thickness and endomyometrial penetration
may reduce the clinical value of such a classification (Kang
et al., 1996). In the meantime it may be helpful to evaluate
results according to the presence or absence of menstrual
pain and the depth of endomyometrial penetration
determined by histology. This would allow comparison of
conservative surgical procedures between centres,
particularly to include those centres that cannot afford
routine use of MRI in the diagnosis of adenomyosis.
If junctional zone hypertrophy is present without
endometrial penetration of the myometrium, it may deserve
a new name, or the definition of adenomyosis could be
changed to include a pre-invasive stage to describe the
junctional zone hypertrophy, adenomyosis, stage 0.
Medical treatment of adenomyosis
There is a paucity of information on the specific effects of
drug therapy on adenomyosis. Drugs used in the treatment of
adenomyosis are mostly the same as those used for
endometriosis, which is easily diagnosed and studied,
whereas often adenomyosis is not diagnosed until after
surgery — either endometrial resection for the treatment of
menorrhagia, or hysterectomy because of the presence of
menorrhagia and dysmenorrhoea in the presence of an
enlarged uterus.
Arguments in favour of medical therapy are the possible
avoidance of surgery and associated complications, such as
adhesions, the limited types of conservative surgery
available, and the tendency of gynaecologists to offer
hysterectomy as the only type of definitive surgery.
Disadvantages of medical therapy are the few reports of
results of treating women with adenomyosis, the commonly
held belief that drug therapy is relatively ineffective in
treating adenomyosis, and drug side-effects.
Women who pursue hysterectomy for adenomyosis have
nearly all been treated previously with drugs for
menorrhagia and/or dysmenorrhoea. These drugs include the
oral contraceptive pill, in continuous mode to avoid
menstruation; progestogens, particularly Provera (medroxy
progesterone acetate), by oral or intramuscular injection; and
less often danazol, gestrinone and GnRH analogues. The
reason the latter three drugs have been used less is that they
are generally prescribed specifically for use in women with
endometriosis, whereas adenomyosis is often not recognized
in young women with menstrual symptoms. The increased
availability of diagnostic methods, MRI, colour Doppler
vaginal ultrasound and uterine biopsy are improving the
recognition of adenomyosis, so that drugs that are specific
for endometriosis may be subject to controlled clinical trial
in the treatment of adenomyosis.
It is difficult to predict the efficacy of drugs reducing
oestrogen or its effect, as comparison of oestrogen receptor
(ER) and progesterone receptor (PR) concentrations in
endometriotic and adenomyotic tissue has not been found in
a literature search. Nevertheless, studies of receptors in
peritoneal, ovarian and rectovaginal endometriosis may have
some relevance to adenomyosis (Nissole et al., 1996). Cyclic
changes are found in both ER and PR content and the
concentration of ER is always lower than in uterine
endometrium, whereas PR content is similar to that in uterine
endometrium. There is hormonal independence of the
endometriotic endometrium as shown by persistence of a
high inactive PR content during the late secretory phase. The
lower level of ER content may be one factor limiting the
effect of drugs which act by lowering or blocking oestrogen
effects. Haemorrhagic and vesicular (blister-like) lesions,
which contain mostly glandular and stromal components,
show evidence of proliferative and secretory change with
menstruation, whereas nodules or papules, which contain
less glands and stroma, and are blue black, brown or white,
show proliferative activity but little or no secretory change
Treatment of adenomyosis
and less or no menstruation (Schweppe, 1996). Only red and
blister lesions, which are better differentiated histologically,
respond to drug therapy (Schweppe, 1996) and such lesions
may be less frequent in adenomyosis.
The results of drug therapy in endometriosis may be most
relevant to women with adenomyosis who have symptoms
of menorrhagia and dysmenorrhoea. Medical therapy aims
to suppress cyclical hormonal changes of ovarian steroid
secretion and inhibit pituitary gonadotrophic secretion or at
least prevent the mid-cycle surge of oestrogen.
Oral contraceptives
Low dose combined oral contraceptives using continuous
therapy, with withdrawal bleeds every 4–6 months, may be
effective in relieving menorrhagia and dysmenorrhoea and
in endometriosis produce equivalent results to oral
contraceptives with higher dose regimens (Moghissi, 1988).
333
and has partly replaced the use of this drug (Fedele et al.,
1989) (Table IV).
Table IV. Side-effects of three antiendometriosis medicines
Side-effect
Danazola
(n = 103)
(%)
Gestrinoneb
(n = 19)
(%)
GnRH (goserelin)a
(n = 204)
(%)
Hot flushes
58
0
98
Acne
54
21
37
Reduced libido
52
–
66
Oily hair/skin
48
16
25
Weight gain
27
42
1
Nausea
12
11
5
Headache
3
–
5
Hirsutism
3
0
0
Voice changes
3
0
Muscle cramps
12
16
0.5
1
Progestogen
Of all the progestogens, oral medroxyprogesterone acetate
(MPA) has been best studied. The therapeutic effects of 30
mg and 50 mg daily are similar; 30 mg daily is associated
with the option of increasing the dose according to clinical
response and bleeding patterns. The major side-effects of
progestogen therapy for endometriosis are breakthrough
bleeding, weight gain, fluid retention (Luciano et al., 1988),
breast tenderness, and mood changes (Mittal and Barwick,
1993).
Danazol
Danazol has been the most commonly used medical
treatment for endometriosis. Three double-blind,
placebo-controlled, prospective trials have randomly
assigned patients to treatment with 100 mg, 200 mg, 400 mg
or 600 mg of danazol (Wingfield and Healy, 1993).
Recurrence rate was lower in those receiving larger doses
(Dmowski et al., 1982). A minimum dosage of 400 mg
danazol per day seems optimal. The androgenic effects of
danazol produce many undesirable side-effects, some of
which may be irreversible, e.g. hirsutism and deepening of
the voice (Table IV).
Gestrinone
Gestrinone is an androgen, a progestogen, an
anti-progestogen, and anti-oestrogen which has been used to
treat endometriosis. It has similar efficacy to danazol (Fedele
et al., 1989). One advantage of gestrinone is its long half-life
when given orally, making twice weekly administration
therapeutic for most patients. The standard dose has been 2.5
mg twice weekly. The side-effects of gestrinone are
predominantly androgenic. It is less androgenic than danazol
aData
from Shaw (1992).
from Fedele et al. (1989).
GnRH = gonadotrophin releasing hormone.
bData
RU486 (mifepristone)
RU486 is a synthetic steroid with antiprogesterone and
antiglucocorticoid activity. It blocks progesterone receptors
in endometrial tissue. Long-term low-dose RU486, 50 mg
daily, achieves anovulation, reducing painful symptoms and
decreasing the extent of endometriosis without an
antiglucocorticoid effect. One study was carried out on 14
women over 6 months of treatment (Kettel, 1996).
Endometriosis scores decreased by ~50%, decreasing pelvic
pain and dysmenorrhoea in all patients. There was no change
in mean serum cortisol.
GnRH agonists
There has been a rapid acceptance of GnRH agonists as
treatment for endometriosis. The drugs are inactive orally
and so must be administered i.v., s.c. (injections or depots) or
via nasal sprays. They have a similar efficacy to danazol
(Shaw, 1992). The side-effects of GnRH agonists are
consequent to the hypo-oestrogenic state induced and are
summarized in Table IV.
Alleviation of pain
Based on currently published studies, it would appear that
GnRH analogues, danazol, gestrinone and MPA show
similar efficacy in terms of laparoscopic resolution of disease
following therapy (Fedele et al., 1997). All studies quoted
showed at least some remission of symptoms in 70–100% of
patients. This compares favourably with placebo arms which
334
C.Wood
show only 18% remission. These effects are most striking for
symptoms of dysmenorrhoea and pelvic pain.
All the current treatment regimens appear to be equally
effective in terms of symptom relief (Wingfield and Healy,
1993). Choice of drug will depend more on side-effects and
cost profiles. While relief of symptoms is effective during
therapy, there is a gradual return of symptoms in some
30–60% of patients by 1 year following therapy.
It is not clear how the information on the effect of drugs on
endometriosis applies to the management of adenomyosis.
There are two considerations. Some patients with
adenomyosis also have endometriosis so the drug effects are
relevant to patient management. More important is the
possibility that women with endometriosis and
dysmenorrhoea or menorrhagia may also have adenomyosis
or junctional zone hypertrophy (Brosens et al., 1995b).
There has been no adequate explanation of why women with
endometriosis outside the uterus have menstrual pain
identical in nature to that which normal women have with
menstruation, i.e. uterine pain. Hysterectomy performed on
women with endometriosis and menstrual symptoms does
not often show endometriosis on the surface or growing into
the uterus from outside. Sometimes adenomyosis is found
after removal of the uterus, and there may be junctional zone
hypertrophy (JZH), which has only recently been
recognized, and may not have been reported by histology. It
seems likely that women with endometriosis and uterine pain
or menorrhagia may have JZH, or experience pain from
prostaglandins released from myometrial endometriotic
tissue. It is probable that the effects of drugs on women with
endometriosis having menstrual symptoms at least partly
reflect the effects of these drugs on adenomyosis or JZH.
Only prospective studies of women with menstrual
symptoms, both with and without endometriosis, having
MRI and uterine biopsy, can elucidate the effects of drugs on
endometriosis, adenomyosis and JZH.
Drugs and adenomyosis
GnRH analogues
A GnRH analogue, leuprolide acetate, has been used to
produce a constant hypo-oestrogenic state in a woman with
histologically proven adenomyosis (Nelson and Corson,
1993). Dysmenorrhoea and desire for conception were the
two complaints. This produced amenorrhoea, control of pain
and uterine shrinkage. Subsequent cyclic use of an oral
contraceptive resulted in recurrence of pain and uterine
growth. Conception occurred after cessation of the GnRH
analogue which was used for 9 months. There have been
other single reports of pregnancy following the use of GnRH
analogues in women with adenomyosis.
Prolonged use of Zoladex for 12 months in women with
fibroids showed persistence of a 7.5% loss of bone mineral
density 1 year after completion of treatment; this loss was not
prevented by coincidental use of medroxyprogesterone
acetate for 9 months (Caird et al., 1997). Danazol has been
used in low dosage, 100 mg/day, to prolong the shrinkage of
fibroids achieved by 6 months of treatment with GnRH
analogues, reducing regrowth by 30% (De Leo et al., 1997).
This was thought to be due to its anti-progesterone effect
which may or may not be relevant to the use of GnRH
analogues in the treatment of adenomyosis.
There is a need for improved diagnosis of adenomyosis so
that controlled trials of the effect of the GnRH analogues on
infertility in adenomyosis can be performed. Numerous
controlled trials of anti-endometriosis drugs in women with
endometriosis have shown no beneficial effect on fertility
when compared to no treatment (Wingfield and Healy,
1993). Nevertheless, GnRH analogues may be used to
control pain and bleeding in the presence of adenomyosis.
Anti-prostaglandins
An increased production of prostaglandin PG12 has been
shown in the tissue of adenomyosis, which is most increased
in women with the most severe dysmenorrhoea (Hoike et al.,
1996). Anti-prostaglandins may be useful in the control of
menstrual pain in adenomyosis. The author uses rectal
indomethacin because of the reduced side-effects and the
possibility of higher myometrial tissue concentration of the
drug compared to oral administration.
Anti-oestrogen, ICI 182 780
Adenomyosis in a pigtailed monkey has been diagnosed by
MRI and treated with a pure anti-oestrogen, ICI 182 780
(Waterton et al., 1993). The pure anti-oestrogen has a high
affinity for the oestradiol receptors but unlike non-steroidal
anti-oestrogens such as tamoxifen, has been shown to be
devoid of partial agonist (oestrogenic) activity. The drug
resulted in a decrease of 87%, 57% and 45% of the
endometrial volume, myometrial volume and lesion width
respectively in the 4 weeks after the second injection.
Oestradiol concentrations remained high. Subsequently
there was further decrease in the myometrial and
endometrial volume but increase in width of the lesion.
Eighty days after treatment began, post-mortem after
euthanasia showed diffuse adenomyosis in the uterus. The
need to kill the monkey instead of performing hysterectomy
is not clear. A pure anti-oestrogen may offer some advantage
Treatment of adenomyosis
in the treatment of adenomyosis and trials are planned to
assess its usefulness in the human.
Topical danazol/progestogen therapy
Adenomyosis has been treated by 200 mg of danazol
contained in an intrauterine device (DIUD) (Igarashi et al.,
1996). Blood danazol levels are undetectable, ovulation was
not inhibited, and side-effects did not occur. The DIUD was
effective in nine of 10 cases in reducing uterine size and
dysmenorrhoea, and pregnancy occurred in three cases after
removal of the DIUD. Another study of the DIUD
containing 300 µg of danazol produced similar results over
6–12 months (Tanoaka et al., 1996). Symptoms improved in
>70% of patients especially for dysmenorrhoea, the DIUD
was shown to be active after 12 months use, mean CA125
concentrations decreased from 295 to 115 U/ml, mean
uterine volume decreased from 369 to 264 cm3 and there
were no changes in liver function or coagulation tests.
Brosens et al. (1996) were unable to duplicate the beneficial
effects of a 106 mg danazol IUD (18 patients) when
comparing this to goserelin (four patients). The patients all
had proven junctional zone hypertrophy on MRI. The lack of
effect of the DIUD may be due to the high expulsion rate, or
to lower endometrial danazol levels resulting from the lower
dose of danazol in the DIUD when compared to the other
studies. Using the DIUD in Australian women, three of four
expelled the DIUD and a larger IUD would be required to
test its efficacy.
The levonorgestrel intrauterine device (LNIUD) has
proven to be effective not only as a contraceptive but also in
the control of menorrhagia. Its action is to produce an
atrophic endometrium. It may be useful in the control of
menorrhagia in the presence of adenomyosis, and possibly
the reduction of dysmenorrhoea. The anti-oestrogenic effect
may reduce the growth of the adenomyotic tissue. A study of
25 women with menorrhagia associated with adenomyosis
diagnosed by vaginal ultrasound has shown that 23 had
relief of menorrhagia persisting for 1 year after use of the
LNIUD (Fedele et al., 1997). Spotting in the first 3 months
was the most common side-effect, one patient asking to
have the device removed because of this. Six patients reported headaches, three breast tenderness, six greasy hair,
seborrhoea or acne, and seven weight gain. Spotting was
well tolerated.
Acknowledgements
I wish to thank Professor David Healy and Mary Wingfield for
assistance in collating data on the use of drug therapy in the
treatment of endometriosis.
335
References
Brosens, J.J. and Barker, F.G. (1995) The role of myometrial needle
biopsies in the diagnosis of adenomyosis. Fertil. Steril., 63,
1347–1349.
Brosens, J.J., de Souza, N.M., Barker, F.G. et al. (1995a) Endovaginal
ultrasonography in the diagnosis of adenomyosis uteri: identifying the
predictive characteristics. Br. J. Obstet. Gynaecol., 102, 471–474.
Brosens, J.J., de Souza, R.N. and Barker, F.G. (1995b) Uterine junctional
zone: function and disease. Lancet, 346, 558–560.
Brosens, J., Briherstaff, H. and de Souza, N. (1996) Intrauterine Danazol vs
Goserelin for the treatment of uterine adenomyosis – a pilot study. Vth
World Congress on Endometriosis, 21–24 October, Yokohama, Japan,
Abstract O-157, p. 81.
Bulletti, C., Polli, V., Negrini, V. et al. (1996) Adhesion formation after
laparoscopy myomectomy. J. Amer. Assoc. Gynecol. Laparosc., 3,
533–536.
Caird, L.E., West, C.P., Lumsden, M.A. et al. (1997)
Medroxyprogesterone acetate with Zoladex for long-term treatment
of fibroids: effects on bone density and patient acceptability. Hum.
Reprod., 12, 436–40.
Chung, Y.P., Lin, H.H., Sheu, B.C. et al. (1997) Adenomyosis in the broad
ligament and tamoxifen: report of a case. J. Obstet. Gynaecol. Res., 23,
69–73.
Cohen, I., Beyth, Y., Tepper, R. et al. (1995) Adenomyosis in
postmenopausal breast cancer patients treated with tamoxifen: a new
entity? Gynecol. Oncol., 58, 86–91.
Darai, E., Dechaud, H., Benifla, J.L. et al. (1997) Fertility after
laparoscopic myomectomy: preliminary results. Hum. Reprod., 12,
1931–1934.
De Leo, V., Morgante, G., Lanzetta, D. et al. (1997) Danazol
administration after gonadotrophin-releasing hormone analogue
reduces rebound of uterine myomas. Hum. Reprod., 12, 357–360.
Dmowski, W.P., Kapetanakis, E. and Scommegna, A. (1982) Variable
effects of danazol on endometriosis at 4 low-dose levels. Obstet.
Gynecol., 59, 408–415.
Diamond, M.P., Daniell, J.F., Feste, J. et al. (1987) Adhesion reformation
and de novo adhesion formation after reproductive pelvic surgery.
Fertil. Steril., 47, 864–866.
Donnez, J. and Nissole, M. (1995) Advanced laparoscopic surgery for the
removal of rectovaginal septum endometriotic or adenomyotic
nodules. Baillière’s Clin. Obstet. Gynaecol., 9, 769–774.
El-Roeiy, A., Dmowski, W.P., Gleicher, N. et al. (1988) Danazol but not
gonadotrophin-releasing hormone agonist suppresses auto-antibodies
in endometriosis. Fertil. Steril., 50, 864–871.
Fedele, L., Bianchi, S., Viezzoli, T. et al. (1989) Gestrinone versus danazol
in the treatment of endometriosis. Fertil. Steril., 51, 781–785.
Fedele, L., Bianchi, S., Dorta, M. et al. (1992) Transvaginal
ultrasonography in the diagnosis of diffuse adenomyosis. Fertil.
Steril., 58, 94–97.
Fedele, L., Bianchi, S., Raffaeli, R. et al. (1997) Treatment of adenomyosis
– associated menorrhagia with a levonorgestrel-releasing intrauterine
device. Fertil. Steril., 68, 426–429.
Goerttler, J. (1930) Myometrial structure during pregnancy. Gegenbaurs
Morphologisches Jahrbuch, 65, 45.
Harkness, M.L.R. and Harkness, R.D. (1959) Changes in the physical
properties of the uterine cervix of the rat during pregnancy. J. Physiol.
(Lond.), 148, 524.
Hirsch, N.A. (1993) Technologies for the Treatment of Menorrhagia and
Uterine Myomas. HealthCare Technology Series No. 10, Australian
Institute of Health & Welfare, April.
Hoike, H., Ikerone, T. and Mori, N. (1996) Relationship between
dysmenorrhoea severity and prostaglandin production in women with
adenomyosis. Vth World Congress on Endometriosis, 21–24 October,
Yokohama, Japan, Abstract P-21, p. 96.
Igarashi, M., Iizuka, M., Ahe, Y. et al. (1996) A new therapy for uterine
adenomyosis. Vth World Congress on Endometriosis, 21–24 October,
Yokohama, Japan, Abstract O-155, p.80.
Jansen, R.P.S. (1991) Prevention of pelvic peritoneal adhesions. Curr.
Opin. Obstet. Gynecol., 3, 369–374.
336
C.Wood
Kang, S., Turner, D.A., Foster, G.S. et al. (1996) Adenomyosis: specificity
of 5mm as the maximum normal uterine junctional zone thickness in
MR images. Am. J. Roentgenol., 166, 1145–1150.
Kettel, L.M. (1996) Treatment of endometriosis with RU486. Vth World
Congress on Endometriosis, 21–24 October, Yokohama, Japan,
Abstract O-136, p. 70.
Lev Gur, M. (1996) The enlarged uterus. Relation of uterine size to
symptoms and histopathologic findings. J. Reprod. Med., 41,
166–170.
Luciano, A.A., Turksoy, R.N. and Carleo, J. (1988) Evaluation of oral
medroxy-progesterone acetate in the treatment of endometriosis.
Obstet. Gynecol., 72, 323–327.
Maher, P. (1995) The place of gasless laparoscopy. Gynaecol. Endosc., 4,
155–158.
Maher, P., Wood, C. and Hill, D. (1995) Excision of endometriosis in the
pouch of Douglas by combined laparovaginal surgery using the Maher
abdominal elevator. JAAGL, 2, 199–202.
McCausland, A.M. (1992) Hysteroscopic myometrial biopsy: its use in
diagnosing adenomyosis and its clinical application. Am. J. Obstet.
Gynecol., 166, 1619–1626.
Mittal, K.R. and Barwick, K.W. (1993) Endometrial adenocarcinoma
involving adenomyosis without true myometrial invasion is
characterized by frequent preceding estrogen therapy, low histologic
grades, and excellent prognosis. Gynecol. Oncol., 49, 197–201.
Moghissi, K.S. (1988) Treatment of endometriosis with
estrogen–progestin combination and progestogens alone. Clin.
Obstet. Gynecol., 31, 823–28.
Nelson, J.R. and Corson, S. (1993) Long term management of
adenomyosis with a gonadotrophin releasing hormone agonist. Fertil.
Steril., 59, 441–443.
Nisolle, M., Donez, J. and Fasanas-Roux, F, (1996) Steroid receptors,
vimentin and cytokeratin expression in endometriotic tissue. Vth
World Congress on Endometriosis, 21–24 October, Yokohama, Japan,
Abstract.
Operative Laparoscopy Study group (1991) Postoperative adhesion
development after operative laparoscopy: evaluation at early
second-look procedures. Fertil. Steril., 55, 700–704.
Pelosi, M.A. and Kadar, N. (1994) Laparoscopically assisted hysterectomy
for uteri weighing 500g or more. J. Am. Assoc. Gynecol. Laparosc., 1,
405–409.
Pelosi, M.A. III and Pelosi, M.A. (1997) Laparoscopic assisted
transvaginal myomectomy. J. Am. Assoc. Gynecol. Laparosc., 4, 241.
Phillips, D.R. (1995) Endomyometrial resection of menopausal women
with annoying uterine bleeding on estrogen replacement therapy. J.
Am. Assoc. Gynecol. Laparosc., 2, 389–393.
Phillips, D.R., Nathanson, H.G., Milim, S.J. et al. (1996) Laparoscopic
bipolar coagulation for the conservative treatment of adenomyomata.
J. Am. Assoc. Gynecol. Laparosc., 4, 19–24.
Popp, L.W., Schwiedessen, J.P. and Gaetje, R. (1993) Myometrial biopsy
in the diagnosis of adenomyosis uteri. Am. J. Obstet. Gynecol., 169,
546–549.
Ravina, J.H., Herbreteau, D., Ciraru-Vigneron, N. et al. (1995) Arterial
embolisation to treat uterine myomata. Lancet, 346, 671–672.
Reinhold, C., McCarthy, S., Bret, P.M. et al. (1996) Diffuse adenomyosis:
comparison of endovaginal US and MR imaging with histopathologic
correlation. Radiology, 199, 151–158.
Schwepppe, K.W. (1996) Therapy of diffuse peritoneal endometriosis. Vth
World Congress on Endometriosis, 21–24 October, Yokohama, Japan,
Abstract O-58, p. 33.
Seinera, P., Arisio, R., Decko, A. et al. (1997) Laparoscopic myomectomy:
indications, surgical technique and complications. Hum. Reprod., 12,
1927–1230.
Shaw, R.W. (1992) An open randomized comparative study of the effect of
goserelin depot and danazol in the treatment of endometriosis. Fertil.
Steril., 58, 265–272.
Stucke, K. (1950) Über das elastische Verhalten der Achillessehne im
Belastungsversuch. Langenbecks Archiv Klin. Chirurg., 265, 579.
Taskin, M., Lallas, T., Shevchuk, M. et al. (1996) P53 expression in
adenomyosis in endometrial carcinoma patients. Gynecol. Oncol., 62,
241–246.
Tanaoka, Y., Slo, R., Yamamoto, Y. et al. (1996) Investigation of treatment
of adenomyosis by intrauterine device that contains danazol. Vth
World Congress on Endometriosis, 21–24 October, Yokohama, Japan,
Abstract O-156, p. 80.
Vercellini, P., Parazzini, F., Oldani, S. et al. (1995) Adenomyosis at
hysterectomy: a study on frequency distribution and patient
characteristics. Hum. Reprod., 10, 1160–1162.
Waterton, J.C., Breen, S.A., Dukes, M. et al. (1993) A case of adenomyosis
in a pigtailed monkey diagnosed by magnetic resonance imaging and
treated with the novel pure antiestrogen, ICI 182 780. Lab. Anim. Sci.,
43, 247–251.
Wingfield, M. and Healy, D. (1993) Endometriosis. In Medical therapy.
Baillière’s Clin. Obstet. Gynaecol., 7, 813–838.
Wood, C. (1960) Rupture of the uterus. Med. J. Aust., 47, 166–169.
Wood, C. (1964) The expansile behaviour of the human uterus. J. Obstet.
Gynaecol Br. Cwlth, 71, 615.
Wood, C. (1992) Indications for endometrial resection. Med. J. Aust., 156,
157–160.
Wood, C. and Maher, P. (1996a) Laparoscopic minilaparotomy: a suitable
addition to major gynaecological surgery. Gynaecol. Endosc., 5,
83–88.
Wood, C. and Maher, P. (1996b) Removal of fibromyomas by laparoscopic
surgery after preoperative GnRH analogue – an alternative to
abdominal hysterectomy or myomectomy. Aust. NZ J. Obstet.
Gynaecol., 36, 464–468.
Wood, C. and Maher, P. (eds) (1997) Hysterectomy. In Laparoscopic
Hysterectomy. Baillière’s Clin. Obstet. Gynaecol., 11, 111–136.
Baillière Tindall, London.
Wood, C. and Rogers, P. (1993) Pregnancy after planned partial
endometrial resection. Aust. NZ J. Obstet. Gynaecol., 33, 316–318.
Wood, C., Maher, P. and Hill, D. (1994) Biopsy diagnosis and conservative
surgical treatment of adenomyosis. J. Am. Assoc. Gynecol. Laparosc.,
1, 313–316.
Yuen, P.M. (1995) Adenomyosis following endometrial rollerball ablation.
Aust. NZ J. Obstet. Gynaecol., 35, 335–366.
Zimmer, F. (1962) Die plastischen Eigenschaften der Uterusmuskulatur.
Geburtsh u. Frauenheilk, 22, 1026.
Received on January 27, 1998; accepted on June 15, 1998
`